How AI Can Help Drivers Avoid Reckless SpeedingThe technology can provide commercial fleet managers with insights into positive driving and identifying opportunities for individual coaching

Over one million people die each year as a result of road traffic crashes. More than 90 per cent of the world's fatalities on the roads occur in low- and middle-income countries, even though these countries have approximately 60 per cent of the world's vehicles. Half of all road traffic deaths are among vulnerable road users: pedestrians, cyclists, and motorcyclists. Road traffic crashes cost most countries three per cent of their gross domestic product.

These numbers by the World Health Organization are scary. What's shocking is that all this can be avoided if people drove responsibly. This is where technology can help. Companies like Nissan, Renault and Hyundai are working towards creating "connected" cars that will provide real-time and contextual navigation which could help drivers ride better.

The Role of Artificial Intelligence

人工智能、人工智能技术发挥了pivotal role in helping develop such cars. The Driveri platform, for instance, by US-based startup Netradyne, which has presence in Asia, with a technology innovation centre in Bengaluru, captures and analyzes almost every second of the driving experience and provides commercial fleet managers a complete view of the activity of their drivers. For example, it can tell whether the driver is maintaining a distance between fellow passengers on the road, driving in proper lanes and following traffic rules.

It is essentially a multi-sensor, cloud-connected, integrated platform and has hardware components, including multiple cameras, LTE, GPS, accelerometer, gyro, and an advanced NVIDIA processor with one TFLOP compute power to aid in Edge Computing.

The company's founder and chief executive officer Avneesh Agrawal says Driveriis a computer vision-based driver safety system. "It helps in identifying and reducing unsafe driving events and also in recognizing and rewarding positive driving performance. The platform was developed to capture every moment of the driving experience, delivering strong value to both the driver and fleet manager," he explains.

What Matters

Netradyne was founded in September 2015 by Agrawal and David Julian. Before this, Agrawal, a Stanford University graduate, was senior vice president of technology and business at chipmaker Qualcomm for more than a decade, while Julian worked as principal engineer in Qualcomm Research for 12 years and also at US space agency NASA's Jet Propulsion Laboratory before starting Netradyne. Last September, the startup raised $21 million funding in Series B funding from investors like Reliance Industries, Microsoft's corporate venture arm M12 and Point72 Ventures.

In December, Netradyne (essentially meaning, "power of vision") partnered with HYUNDAI MNSOFT, a Hyundai Motor Group subsidiary specializing in location-based services (LBS), to develop HD maps to drive forward their future vehicle technologies.

The reason for starting Netradyne, says Agrawal, was "computation power was increasing exponentially, connectivity was ubiquitously available and getting cheaper by the day, and AI algorithms were maturing. We saw a strong case for IoT devices with Edge computing to be deployed for road safety."

"For example, WHO's Global status report on road safety 2015, reflecting information from 180 countries, indicated that worldwide the total number of road traffic deaths stood at 1.25 million per year. Our aim was to address this issue with the help of AI," he adds.